KR102564218B1 - System for predicting flooding using drone images and artificial intelligence - Google Patents

Abstract

The present invention relates to an inundation flood prediction system, and in particular, an ortho-photo-graph and a digital elevation model based on an image taken by a drone that can reflect the recent environment as it is and provides accurate numerical data. (Digital Elevation Model: DEM) and digital surface model (DSM) are provided with a flooding forecasting system that can visually express the area and degree of flooding, and the visualization information of the flooded area is It can be used for identifying areas vulnerable to inundation and rapid disaster response, and can provide objective numerical data information in selecting the installation location of safety facilities in preparation for floods, such as installation of infiltration blocks and storage tanks.
In addition, it is possible to support work by outputting drawings and visual simulation values necessary for various reports, documents, and data creation, and in the long term, it is possible to build and utilize a DB in terms of infrastructure maintenance through continuous drone image learning. have

Description

System for predicting flooding using drone images and artificial intelligence {System for predicting flooding using drone images and artificial intelligence}

The present invention relates to an inundation flood prediction system, and in particular, an ortho-photo-graph and a digital elevation model based on an image taken by a drone that can reflect the recent environment as it is and provides accurate numerical data. It is about a flood flood prediction system using drone images and artificial intelligence that can visually express flood flood area and degree in (Digital Elevation Model: DEM) and Digital Surface Model (DSM).

In general, as the frequency of floods increases due to recent climate change and global warming, a function that can more accurately predict and visualize inundation flooding situations is required. Based on digital topographical maps and topographical information, it is the extent to which the propagation path of floods is identified through 2D inundation flooding and hydrodynamic analysis based on Geographic Information System (GIS).

However, there are no grades according to the degree of permeability of water according to the land cover map, and the land cover also has problems that are not realistic when compared to the actual environment.

In other words, in the case of Japan, the type of land cover is subdivided and the permeability is applied. In the case of Korea, the existing inundation-related prediction method uses the land cover map provided by the "Environmental Spatial Information Service", and the permeability is 0% or Because it is classified as 100%.

In addition, since the land cover boundary is not updated according to the land development status, greening project, and safety facility installation, there is a limitation in deriving accurate simulation results.

An object of the present invention to solve the above-mentioned problems is related to a flooding flood prediction system, in particular, based on an image taken by a drone that can reflect the recent environment as it is and provides accurate numerical data, Ortho-photo-graph), digital elevation model (DEM), and digital surface model (DSM) that can visually express the area and degree of inundation and inundation using drone images and artificial intelligence. It is to provide a predictive system.

In order to achieve the above object, the feature of the inundation and flooding prediction system using drone images and artificial intelligence according to the present invention is that, in the prediction system for preparing for flooding and flooding, information on land cover maps and drones from public institutions is provided. an input unit that receives image data captured through aerial photography, generates orthoimages and DSM/DEM data based on the received image data, and then outputs a land cover map, orthoimages, and DEM/DSM data; After receiving the land cover map, orthoimage, and DSM/DEM data output from the input unit to predict inundation and inundation based on the land cover map or orthoimage, the land cover boundary of the region is set, and then the information on the amount of precipitation in the region is used as the basis. A calculation unit composed of a deep learning object detection algorithm that performs a simulation according to the permeability according to the type of land cover and the altitude value of the DSM, and learns information about the type of land cover in the orthographic image and the permeability of that type of land cover is to include

As an additional feature of the inundation and flooding prediction system using drone images and artificial intelligence according to the present invention to achieve the above object, the input unit provides a land cover map for an area where flood damage is expected to occur from the corresponding public institution. a land cover map input module that receives an input; A drone shooting data input module that receives aerial image data of the current state of the flood damage expected area taken by the drone; an orthoimage conversion module for converting the aerial image data acquired through the drone shooting data input module into an orthoimage; And a DSM / DEM information acquisition module for acquiring DSM information generated in the process of converting aerial image data into orthoimages through the orthoimage conversion module and acquiring DEM information obtained by processing the corresponding DSM information according to a request. there is

As an additional feature of the inundation and inundation prediction system using drone image and artificial intelligence according to the present invention for achieving the above object, the calculation unit includes land cover map, orthoimage, and DEM and DSM information obtained through the input unit. Input with a deep learning object detection algorithm that sets the boundary of land cover based on a land cover map or an orthographic image, and learns information about the type of land cover in the orthographic image and the permeability of that type of land cover. a data complex processing module; a precipitation input module for receiving precipitation (minimum value, average value, history data according to maximum value) according to characteristics of expected or simulated flood damage occurrence area; a flooding/inundation simulation module for performing a simulation according to the altitude value of the DSM and permeability according to the land cover type according to the precipitation input module based on the data pre-processed by the input data complex processing module; and an information generation module for generating a result of the simulation performed by the inundation/inundation simulation module in a TIFF format such as DSM and a GIF file.

As another additional feature of the inundation and inundation prediction system using drone images and artificial intelligence according to the present invention for achieving the above object, the input data composite processing module includes land cover information input through the input unit and An input data complex operation module that receives orthoimages acquired through images taken by drones, corresponding DSM information, and DEM information according to the DSM information and matches corresponding information of the same region; an orthoimage standard boundary setting module for reorganizing the information matched by the input data complex operation module around an orthoimage to set a land cover boundary; and a land cover standard boundary setting module for setting the boundary of the land cover by reorganizing the information matched by the input data complex calculation module around the land cover map.

As another additional feature of the inundation flood prediction system using drone image and artificial intelligence according to the present invention to achieve the above object, it is interlocked with the input data complex calculation module, and the land cover in the orthoimage is It is to further include an artificial intelligence deep learning module consisting of a deep learning object detection algorithm that is a type and learns information about land cover permeability of that type, thereby increasing the efficiency of the input data complex calculation module.

As another additional feature of the inundation/inundation prediction system using drone images and artificial intelligence according to the present invention for achieving the above object, the inundation/inundation simulation module is pre-processed in the input data complex processing module. a flood simulation module for receiving data and simulating whether or not a flood situation occurs based on water permeability according to land cover types according to the precipitation input module; and an overflow simulation module for simulating whether an overflow situation occurs by receiving data pre-processed by the input data composite processing module and applying precipitation data according to the precipitation input module based on the altitude value of the DSM.

Another additional feature of the inundation flood prediction system using drone images and artificial intelligence according to the present invention to achieve the above object is to change the location of the surrounding environment or disaster prevention facilities based on the simulation performed by the calculation unit It is to further include an analysis unit that performs a simulation according to the corresponding change.

As another additional feature of the inundation and overflow prediction system using drone image and artificial intelligence according to the present invention to achieve the above object, the analysis unit is based on the simulation performed in the flood / flood simulation module of the operation unit. a simulation information analysis module that analyzes simulation results; It is to include an environment variable simulation module that performs a simulation according to the change by varying the location of the surrounding environment or disaster prevention facility, etc. based on the simulation performed by the flood / overflow simulation module of the calculation unit.

As another additional feature of the inundation overflow prediction system using drone images and artificial intelligence according to the present invention for achieving the above object, the environment variable simulation module is based on the information analyzed by the simulation information analysis module A cross-section generating module for generating a cross-sectional view of the simulation area; A penetration block simulation information variable module for generating information for simulation by varying the installation position of the penetration block based on the cross section information of the land surface generated through the cross section generation module; a storage tank simulation information variable module generating information for simulation by varying the installation position and capacity of the storage tank based on the cross-sectional information of the land surface generated through the cross-section generating module; And re-performing the simulation of inundation and flooding based on the information of the seepage block and the storage tank, the information of which is changed through the penetration block simulation information variable module and the storage tank simulation information variable module, to obtain information about changes in the environment for disaster prevention. It is to include a variable information simulation module for

If such a flood flooding prediction system according to the present invention is provided, the visualization information of the flooded area can be used for identifying areas vulnerable to flooding and flooding in real-time national land management and for prompt disaster response work, and safety facilities against floods such as installation of infiltration blocks and storage tanks Objective numerical data information can be provided in selecting the installation location.

In addition, it is possible to support work by outputting drawings and visual simulation values necessary for various reports, documents, and data creation, and in the long term, it is possible to build and utilize a DB in terms of infrastructure maintenance through continuous drone image learning. have

1 is an exemplary block configuration diagram of a flood inundation prediction system using drone images and artificial intelligence according to the present invention
2 to 4 are examples of photos of land cover maps, orthographic images, and DSMs used in the inundation and flooding prediction system using drone images and artificial intelligence according to the present invention.
5 to 7 are detailed configuration examples of main components of the inundation overflow prediction system using drone images and artificial intelligence according to the present invention

Hereinafter, prior to describing the inundation overflow prediction system using drone image and artificial intelligence according to the present invention with reference to the accompanying drawings, the terms used in the present invention are summarized, and the digital elevation model (Digital Elevation Model: DEM) is a numerical model that expresses a bare earth part excluding buildings, trees, and artificial structures among real-world topographic information.

In addition, a digital surface model (hereinafter referred to as DSM) is a model expressing all information in the real world, that is, topography, trees, buildings, artificial structures, and the like.

In addition, ortho-photo-graph is an image obtained by correcting geometrical distortion due to topographical relief such as height difference or inclination of image information such as aerial photographs or artificial satellites, and converting all objects into images when looking down vertically. It refers to an image map in which coordinates and cycles are written and aggregated according to a certain standard.

Referring to attached Figure 1, looking at the configuration of the inundation flood prediction system using drone image and artificial intelligence according to the present invention, information on land cover map from public institutions and image data photographed from the air through drones are input and based on this An input unit 100 that generates orthoimages and DSM/DEM data and then outputs the land cover map, orthoimages, and DEM/DSM data, and the land cover map, orthoimages, and DSM/DEM data output from the input unit 100 After setting the land cover boundary of the region to predict inundation and flooding based on the land cover map or orthographic image, simulation according to the altitude value of the DSM and permeability according to the type of land cover based on the information on the amount of precipitation in the region based on the calculation unit 200 consisting of a deep learning object detection algorithm that learns information about the type of land cover in the orthoimage and the permeability of that type of land cover, and the simulation performed by the calculation unit 200. An analysis unit 300 that performs a simulation according to the change by varying the surrounding environment or the location of a disaster prevention facility, and the result of each simulation performed through the operation unit 200 and the analysis unit 300 and the corresponding area It is largely composed of an output unit 400 that expresses information.

Looking at the details of the above-described configuration, the input unit 100 includes a land cover map input module 110 that receives a land cover map for an area where flood damage is expected to occur from a corresponding public institution, and the corresponding flood water photographed through a drone. A drone shooting data input module 130 that receives aerial image data of the current state for an area where damage is expected to occur, and an ortho image conversion that converts the aerial image data acquired through the drone shooting data input module 130 into an ortho image The module 150 and the orthoimage conversion module 150 acquire DSM information generated in the process of converting aerial image data into orthoimages through the orthoimage conversion module 150, and acquire DEM information by processing the corresponding DSM information according to a request. /DEM information acquisition module 160.

At this time, the image of the land cover map obtained through the land cover map input module 110 is as shown in FIG. 2, and the image of the ortho image generated through the ortho image conversion module 150 is shown in FIG. 3. As shown, the DSM image obtained through the DSM/DEM information acquisition module 160 is as shown in FIG. 4 attached.

In addition, the calculation unit 200 receives the land cover map, orthoimage, and DEM and DSM information obtained through the input unit 100 and sets the boundary of land cover based on the land cover map or orthoimage, and within the orthoimage An input data complex processing module 210 having a deep learning object detection algorithm that learns information about the type of land cover and the permeability of that type of land cover, and the amount of precipitation (minimum value , average value, history data according to the maximum value) and the land cover type according to the precipitation input module 220 based on the pre-processed data in the input data composite processing module 210 and the precipitation input module 220 receiving input. A flood / flood simulation module 230 that performs simulations according to the permeability and the altitude value of the DSM according to the permeability, and the result of the simulation performed by the flood / flood simulation module 230 in a TIFF format and GIF file, such as DSM. Generating It includes an information generating module 250.

At this time, the input data complex processing module 210, as shown in the attached FIG. 5, the land cover information input through the input unit 100 and the orthophoto acquired through the drone's image, and the corresponding DSM information, and an input data composite calculation module 211 that receives DEM information according to the DSM information and matches information of the same region, and reorganizes the matched information in the input data composite calculation module 211 around an orthoimage. The land cover standard boundary setting module 212 for setting the land cover boundary and the information matched in the input data compound calculation module 211 are reorganized around the land cover map to set the land cover boundary. It consists of modules 213.

At this time, an artificial intelligence deep learning module 211A interlocking with the input data compound calculation module 211 is further provided. It is composed of a deep learning object detection algorithm that learns information on land cover permeability, and is intended to increase the efficiency of the input data complex calculation module 211.

As shown in FIG. 6, the inundation/inundation simulation module 230 receives the data pre-processed by the input data composite processing module 210 and calculates the permeability according to the land cover type according to the precipitation input module. The flood simulation module 231, which simulates whether or not a flood situation occurs as a standard, and the data pre-processed by the input data complex processing module 210 are input and entered into the precipitation input module 220 based on the altitude value of the DSM. It is composed of an overflow simulation module 232 that simulates whether an overflow situation occurs by applying the corresponding precipitation data.

In addition, the analysis unit 300 includes a simulation information analysis module 310 that analyzes simulation results based on the simulation performed in the inundation/inundation simulation module 230 of the operation unit 200, and the operation unit 200 Based on the simulation performed by the inundation/inundation simulation module 230 of the environment variable simulation module 320 that performs simulation according to the change by varying the location of the surrounding environment or disaster prevention facilities.

At this time, as shown in FIG. 7, the environment variable simulation module 320 includes a cross-section generation module 321 for generating a cross-sectional view of the simulation area based on the information analyzed by the simulation information analysis module 310, and A penetration block simulation information variable module 322 for generating information for simulation by varying the installation position of the penetration block based on the cross section information of the surface generated through the cross section generation module 321, and the cross section generation module ( 321), a storage tank simulation information variable module 323 for generating information for simulation by varying the installation position and capacity of the storage tank based on the cross-sectional information of the land surface, and the penetration block simulation information variable module 322, A variable information simulation module for obtaining information on changes in the environment for disaster prevention by re-performing the simulation of inundation and flooding based on the information of the infiltration block and the storage tank whose information is changed through the storage tank simulation information variable module 323 ( 324).

Looking at the overall operation and operation method of the inundation overflow prediction system according to the present invention configured as described above, the first part to be looked at is the part related to the land cover boundary.

For the inundation and flooding simulation according to the present invention, boundary treatment of land cover is important, and boundary treatment can be selected and applied by the user based on land cover map or orthographic image according to business purpose and priority of information. .

First of all, if the simulation is performed centering on the land cover map, the existing land cover boundary is maintained and the information properties of the shapefile (hereinafter referred to as SHP) are mapped.

In this case, the SHP format means that vector features such as points, lines, and polygons representing wells, rivers, and lakes can be spatially described, each of which has an attribute that describes the item, typically a name or a temperature. It is a geospatial vector data format for information systems (GIS) software.

Therefore, if there are various land covers in one area, the simulation is performed with the type of cover that occupies the largest area and the permeability.

On the other hand, if the simulation is performed centering on the orthoimage, it is necessary to perform the simulation by newly defining the spatial information according to the different types of land cover in the actual image, ignoring the boundary of the land cover. Although it has the disadvantage of increasing time and resources (computer performance, etc.), it has the advantage of being able to simulate more realistically than simulations centered on land cover created in the past because it can be simulated at the present time. Therefore, the land cover type and permeability can be modified by the user.

In the case of Korea, using the environmental spatial information service, information on the land cover map can be provided as an SHP file. However, since the information on the land cover map is updated based on a certain point in time (the time of survey work by a public institution), it is not data at the current point in time.

At this time, the drone includes data such as altitude, coordinates, and shooting angle as metadata in the image data of the photo.

Accordingly, the land cover map input module 110 of the input unit 100 receives input of a land cover map for an area where flood damage is expected to occur from a corresponding public institution through the Internet environment.

At this time, the image according to the land cover map has a form as shown in FIG. 2 attached.

On the other hand, the aerial image data of the current state for the flood damage expected area taken by the drone is provided to the drone shooting data input module 130 through the memory device, and output from the drone shooting data input module 130 Based on the metadata included in the image of the flood damage expected area, the ortho image conversion module 150 converts the image into an ortho image as shown in attached FIG. 3 and outputs it.

At this time, DSM information generated in the process of obtaining an orthoimage in the orthoimage conversion module 150 is acquired through the DSM/DEM information acquisition module 160, and a DSM information image as shown in FIG. 4 is generated. .

At this time, when a request (control) that DEM information is required comes in, the DSM/DEM information acquisition module 160 generates DEM information based on the acquired DSM information. In addition, it is also possible to acquire DEM information disclosed from the outside.

In general, depending on the user's purpose of use, either DSM or DEM is input. In general, DSM is used in urban areas with many buildings and DEM is used in forested and mountainous areas with many trees.

Thereafter, in the input data complex processing module 210 of the calculation unit 200, the land cover map and ortho image obtained through the land cover map input module 110, the ortho image conversion module 150, and the DSM/DEM information acquisition module 160 After receiving DSM/DEM information and matching the information of the same area, the user selects the boundary of land cover based on the land cover map or the boundary of land cover based on the orthographic image.

The operation of the input data composite processing module 210 described above includes the input data composite calculation module 211, the orthoimage standard boundary setting module 212, the land cover map standard boundary setting module 213, and the artificial intelligence deep learning module. 211A, and the operation of each component is omitted because it overlaps with the above-described configuration description of FIG. 5 .

When the land cover boundary is set through either the orthoimage reference boundary setting module 212 or the land cover reference boundary setting module 213 through the input data complex processing module 210, the flood/inundation simulation module 230 In , based on the amount of precipitation (history data according to the minimum, average, and maximum values) of the corresponding region input through the precipitation amount input module 220, whether or not flooding or flooding occurs and the degree thereof is simulated.

At this time, the result of the simulation performed by the inundation/inundation simulation module 230 is generated in a TIFF format such as DSM and a GIF (dynamic image) in the information generating module 250. The information generating module 250 will be described later. It also processes the simulation result of the analysis unit 300 and performs the processing function of the information expressed through the output unit 400.

In addition, as shown in FIG. 6, the inundation/inundation simulation module 230 may be divided into a flood simulation module 231 and an inundation simulation module 232, and flood damage and flood damage are calculated through one simulation. can all be expected.

In this way, when the flooding simulation for any specific area is performed, it is necessary to check the location and capacity of disaster prevention facilities to prevent flooding or flooding based on this, and this is done through the analysis unit 300.

That is, the simulation information analysis module 310 of the analysis unit 300 analyzes the results of the simulation based on the simulation performed in the inundation/inundation simulation module 230 of the calculation unit 200, such as vulnerable districts and vulnerable types, etc. read

Thereafter, in the environment variable simulation module 320, based on the simulation performed in the flood / overflow simulation module 230 of the calculation unit 200, while varying the location and capacity of the disaster prevention facility, it is simulated whether flooding or flooding is prevented. will read

Therefore, it is possible to check the optimal environment (installation location of the infiltration block, installation location and capacity of the storage tank) for preventing inundation or flooding through the environment variable simulation module 320.

A description of the cross-section generating module 321, the variable penetration block simulation information module 322, the variable storage tank simulation information module 323, and the variable information simulation module 324 constituting the variable environment simulation module 320 is shown in the foregoing figure. Since it overlaps with the description of 7, the description is omitted.

Through the inundation and overflow prediction system according to the present invention described above, the existing inundation and overflow prediction system could not smoothly change the infiltration block, land cover change, storage tank installation, etc. virtually, but in the case of this prediction system, the installation target of the infiltration block (Area) can be easily changed and input by the user, and after inputting the size (width x length x height, capacity) of the storage tank, the user can place (x, y, z) at the desired location to perform simulation.

Therefore, according to the current disaster safety facility installation for public safety, it has an effect of facilitating the derivation of optimal alternatives in areas prone to flooding and flooding, such as frequent flooding and localized heavy rains.

Although the preferred embodiments of the present invention have been shown and described above, the present invention is not limited to the specific embodiments described above, and without departing from the gist of the present invention claimed in the claims, in the technical field to which the method belongs Various modified implementations are possible by those skilled in the art, and these modified implementations should not be individually understood from the technical spirit or perspective of the present invention.

100: input unit
110: land cover map input module
130: Drone shooting data input module
150: ortho image conversion module
160: DSM/DEM information acquisition module
200: calculation unit
210: input data complex processing module
211: input data complex calculation module
211A: Artificial Intelligence Deep Learning Module
212: ortho image reference boundary setting module
213: land cover map standard boundary setting module
220: precipitation input module
230: inundation/flood simulation module
231: flood simulation module
232: overflow simulation module
240: operation control module
250: information generating module
300: analysis unit
310: simulation information analysis module
320: environment variable simulation module
321: section generation module
322: penetration block simulation information variable module
323: storage tank simulation information variable module
324: variable information simulation module
400: output unit

Claims (9)

In the prediction system for preparing for inundation flooding,
An input unit that receives information on land cover maps from public institutions and image data captured by drones, generates ortho images and DSM/DEM data based on them, and then outputs the land cover maps, ortho images, and DEM/DSM data;
After receiving the land cover map, orthoimage, and DSM/DEM data output from the input unit to predict inundation and inundation based on the land cover map or orthoimage, the land cover boundary of the region is set, and then the information on the amount of precipitation in the region is used as the basis. A calculation unit composed of a deep learning object detection algorithm that performs a simulation according to the permeability according to the type of land cover and the altitude value of the DSM, and learns information about the type of land cover in the orthographic image and the permeability of that type of land cover including,
The calculation unit,
The land cover map, orthoimage, and DEM and DSM information obtained through the input unit are input, and the boundary of land cover is set based on the land cover map or the orthoimage, and the type of land cover in the orthoimage and the permeability of the land cover of that type are set. an input data composite processing module having a deep learning object detection algorithm that learns about information about;
a precipitation input module for receiving precipitation (minimum value, average value, history data according to maximum value) according to characteristics of expected or simulated flood damage occurrence area;
a flooding/inundation simulation module for performing a simulation according to the altitude value of the DSM and permeability according to the land cover type according to the precipitation input module based on the data pre-processed by the input data complex processing module; and
An information generation module for generating the result of the simulation performed in the inundation / overflow simulation module in a TIFF format such as DSM and a GIF file,
An analysis unit for performing a simulation according to the change by varying the location of the surrounding environment or disaster prevention facility based on the simulation performed by the calculation unit;
The analysis unit analyzes simulation results based on the simulation performed by the inundation/inundation simulation module of the calculation unit and reads a vulnerable district and a vulnerable type; a simulation information analysis module;
Based on the simulation performed by the inundation / overflow simulation module of the calculation unit, the drone image and artificial Inundation flood prediction system using intelligence.
According to claim 1,
The input unit,
a land cover map input module for receiving input of a land cover map for an area expected to have flood damage from a corresponding public institution;
A drone shooting data input module that receives aerial image data of the current state of the flood damage expected area taken by the drone;
an orthoimage conversion module for converting the aerial image data acquired through the drone shooting data input module into an orthoimage; and
Including a DSM/DEM information acquisition module that acquires DSM information generated in the process of converting aerial image data into orthoimages through the orthoimage conversion module and acquires DSM information obtained by processing the corresponding DSM information as required. Inundation overflow prediction system using drone images and artificial intelligence.
delete According to claim 1,
The input data complex processing module,
Input data composite calculation for receiving the land cover information input through the input unit, the ortho image obtained through the image taken by the drone, the corresponding DSM information, and the DEM information according to the DSM information and matching the corresponding information of the same region. module;
an orthoimage standard boundary setting module for reorganizing the information matched by the input data complex operation module around an orthoimage to set a land cover boundary; and
Inundation flood prediction system using drone image and artificial intelligence, characterized in that it comprises a land cover reference boundary setting module for setting the land cover boundary by reorganizing the information matched in the input data complex calculation module around the land cover map.
According to claim 4,
An artificial intelligence deep learning module consisting of a deep learning object detection algorithm that interworks with the input data complex calculation module and learns information about the type of land cover in the orthoimage and the permeability of that type of land cover Further comprising, Flood flooding prediction system using drone image and artificial intelligence, characterized in that for increasing the efficiency of the input data complex calculation module.
According to claim 1,
The inundation / overflow simulation module,
a flood simulation module for receiving the data pre-processed by the input data complex processing module and simulating whether or not a flood situation occurs based on permeability according to land cover types according to the precipitation input module;
A flood simulation module for simulating whether a flood situation occurs by applying the precipitation data according to the precipitation input module based on the altitude value of the DSM by receiving the data pre-processed by the input data composite processing module. Inundation overflow prediction system using drone image and artificial intelligence.
delete delete According to claim 1,
The environment variable simulation module includes a cross-section generating module for generating a cross-sectional view of the simulated area based on the information analyzed by the simulation information analysis module;
A penetration block simulation information variable module for generating information for simulation by varying the installation position of the penetration block based on the cross section information of the land surface generated through the cross section generation module;
a storage tank simulation information variable module generating information for simulation by varying the installation position and capacity of the storage tank based on the cross-sectional information of the land surface generated through the cross-section generating module; and
To obtain information on changes in the environment for disaster prevention by re-simulating inundation and flooding based on the information of the seepage block and storage tank, the information of which is changed through the penetration block simulation information variable module and the storage tank simulation information variable module Flood flooding prediction system using drone image and artificial intelligence, characterized in that it includes a variable information simulation module.

Images